During Release 12, the Long Term Evolution (LTE) standard has been extended with support of device-to-device (D2D) (specified as “sidelink”) features targeting both commercial and public safety applications. An example application enabled by Release 12 LTE is device discovery, where devices are able to sense the proximity of another device and associated application by broadcasting and detecting discovery messages that carry device and application identities. Another example application enabled by Release 12 LTE is direct communication based on physical channels terminated directly between devices.
One of the potential extensions for D2D is support of vehicle-to-anything-you-can-imagine (V2x) communication. V2x communication includes any combination of direct communication between vehicles, pedestrians and infrastructure. V2x communication may take advantage of a network infrastructure, when available, but at least basic V2x connectivity should be possible even in case of lack of coverage. Providing an LTE-based V2x interface may be economically advantageous because of the LTE economies of scale, and it may enable tighter integration between communications with the network infrastructure (V2I), vehicle-to-pedestrian (V2P) communications, and vehicle-to-vehicle (V2V) communications, as compared to using a dedicated V2x technology.
V2x communications may carry both non-safety and safety information, where each of the applications and services may be associated with a specific set of requirements (e.g., in terms of latency, reliability, capacity, etc.). For example, the European Telecommunications Standards Institute (ETSI) has defined two types of messages for road safety: the Cooperative Awareness Message (CAM) and the Decentralized Environmental Notification Message (DENM).
The CAM message is intended to enable vehicles, including emergency vehicles, to notify their presence and other relevant parameters in a broadcast fashion. These messages target other vehicles, pedestrians, and infrastructure, and are handled by their applications. CAM messages also serve as active assistance to safety driving for normal traffic. The availability of a CAM message is indicatively checked for every 100 ms, yielding a maximum detection latency requirement of less than or equal to 100 ms for most messages. The latency requirement for a pre-crash sensing warning, however, is 50 ms.
The DENM message is event-triggered, such as by braking. The availability of a DENM message is also checked for every 100 ms. The requirement of maximum latency is less than or equal to 100 ms.
The package size of CAM and DENM messages varies from 100+ to 800+ bytes, and the typical size is around 300 bytes. The message is supposed to be detected by all vehicles in proximity.
The Society of the Automotive Engineers (SAE) has also defined the Basic Safety Message (BSM) for Dedicated Short Range Communications (DSRC), with various message sizes defined. According to the importance and urgency of the messages, BSMs are further classified into different priorities.
The 3rd Generation Partnership Project (3GPP) is discussing enhancements to the LTE physical layer in order to improve the efficiency of V2x communication. This includes communications between devices (e.g., vehicles) as well as communications between devices and the network infrastructure (e.g., roadside units). The evolution of the sidelink physical layer format is considered by 3GPP as the starting point for such enhancements. One of the key enhancement areas is the design of demodulation reference signal (DMRS) to allow low error probability when detecting signals transmitted by devices in high mobility. One of the prioritized deployment scenarios assumes a carrier frequency around 6 GHz.
The sidelink physical layer format was designed assuming lower carrier frequency and lower relative mobility between nodes, which results in lower Doppler spread between nodes. One of the approaches currently studied by RANI for coping with increased Doppler spread consists of densifying the time-density of the DMRS pattern in each subframe, which reduces the need to interpolate/extrapolate channel samples between DMRS transmission instances.